Research in context
Evidence before this study
We searched PubMed and MEDLINE for clinical trials published in English between Jan 1, 2000, and July 1, 2016, using the terms “high-flow oxygen”, “infants”, “children”, “p*ediatrics”, “bronchiolitis”, but could not find randomised trial evidence for the use of high-flow oxygen compared to standard low-flow nasal cannula oxygen. We identified a randomised trial (n=19) comparing variable flow oxygen (4–8 L/min) with headbox oxygen in bronchiolitis through Beggs and colleagues' Cochrane review, whereas Mayfield and colleagues' review examining high-flow oxygen in any paediatric context other than bronchiolitis found none. A recent Bangladeshi trial (n=225) compared CPAP, 2 L/kg per min high-flow oxygen, and standard therapy in children with severe pneumonia, but was stopped early because of the significant benefit of CPAP over standard therapy for mortality (three deaths vs ten deaths). Paediatric observational studies and audits suggest that high-flow oxygen is beneficial, and Manley and colleagues were able to show equivalence to CPAP post-extubation of neonates. However, other studies in neonates and adults have found no difference in their primary outcome.
Added value of this study
To our knowledge, this study is the first randomised trial to examine HFWHO in a paediatric, ward-based cohort. HFWHO and standard therapy were both effective when standardised approaches to starting and weaning oxygen were used, and early use of HFWHO did not alter the overall course of the bronchiolitis. HFWHO prevented clinical deterioration in significantly more infants than standard therapy and was able to reverse deterioration in 63% (95% CI 45–77) of the 32% (23–41) of those who were not adequately supported by standard therapy.
Implications of all the available evidence
For children with moderate bronchiolitis, HFWHO is safe and effective at 1 L/kg per min (maximum FiO12 0·6). For children who are not supported by standard therapy, HFWHO might reduce the need for admission to ICU with substantial cost savings. Whereas observational studies have suggested that HFWHO is more beneficial compared with standard therapy, our study has not shown superiority for the primary endpoint of time on oxygen. Future research is required to test HFWHO as a rescue treatment in bronchiolitis, to test its application to other paediatric respiratory conditions, and to examine the safety and effectiveness of standardised procedures for starting and weaning of oxygen with economic outcomes.
CPAP=continuous positive airway pressure. HFWHO=high-flow warm humidified oxygen.